Indicate the one major type of limitation that is possible in each of the situations described below. Choose from the following: selection bias, information bias, confounding, random error, or no error. Choose only one per example.
1) A population survey of the United States is conducted, where a probability sample is drawn, such that for every individual in the U. S., the probability of their inclusion in the study is calculable.
2) In a study of BMI as a possible risk factor for breast cancer, it was discovered that in general the cases with breast cancer were of higher socioeconomic status than the controls without breast cancer, and that high BMI was more common among those of low socioeconomic status than their higher SES counterparts. Initially, before the role of socioeconomic status was taken into account, contrary to the expectations of the investigators, low BMI was found to be associated with breast cancer. However, stratified analyses within high and low socioeconomic groups revealed that, in fact, if socioeconomic status is held constant, high BMI is associated with breast cancer.
3) Physicians may examine women who use oral contraceptives more often or more thoroughly than women who do not. An association is observed between phlebitis and oral contraceptive use.
4) A stratified analysis of the association between depression and obesity was conducted within three racial/ethnic groups: Non-Hispanic Whites, African Americans and Hispanics. Unfortunately, the numbers of participants in each of the minority groups was between a third and half that of the Non-Hispanic Whites. The results were as follows: Non-Hispanic Whites: odds ratio: 2.02 (95% confidence interval: 1.67-3.58); African Americans: odds ratio: 2.14 (95% confidence interval: 0.89-5.08); Hispanics: odds ratio: 1.98 (95% confidence interval: 0.67-8.26). The investigators concluded that for each racial/ethnic group, the odds of being depressed among the obese was double that of the non-obese.